Cody Hamilton

Phase II study of belagenpumatucel-L, a transforming growth factor beta-2 antisense gene-modified allogeneic tumor cell vaccine in non-small-cell lung cancer.

PURPOSE Belagenpumatucel-L is a nonviral gene-based allogeneic tumor cell vaccine that demonstrates enhancement of tumor antigen recognition as a result of transforming growth factor beta-2 inhibition. PATIENTS AND METHODS We performed a randomized, dose-variable, phase II trial involving stages II, IIIA, IIIB, and IV non-small-cell lung cancer patients. Each patient received one of three doses (1.25, 2.5, or 5.0 x 10(7) cells/injection) of belagenpumatucel-L on a monthly or every other month schedule to a maximum of 16 injections. Immune function, safety, and anticancer activity were monitored. RESULTS Seventy-five patients (two stage II, 12 stage IIIA, 15 stage IIIB, and 46 stage IV patients) received a total of 550 vaccinations. No significant adverse events were observed. A dose-related survival difference was demonstrated in patients who received > or = 2.5 x 10(7) cells/injection (P = .0069). Focusing on the 61 late-stage (IIIB and IV) assessable patients, a 15% partial response rate was achieved. The estimated probabilities of surviving 1 and 2 years were 68% and 52%, respectively for the higher dose groups combined and 39% and 20%, respectively, for the low-dose group. Immune function was explored in the 61 advanced-stage (IIIB and IV) patients. Increased cytokine production (at week 12 compared with patients with progressive disease) was observed among clinical responders (interferon gamma, P = .006; interleukin [IL] -6, P = .004; IL-4, P = .007), who also displayed an elevated antibody-mediated response to vaccine HLAs (P = .014). Furthermore, positive enzyme-linked immunospot reactions to belagenpumatucel-L showed a correlation trend (P = .086) with clinical responsiveness in patients achieving stable disease or better. CONCLUSION Belagenpumatucel-L is well tolerated, and the survival advantage justifies further phase III evaluation.

New-Onset Postoperative Atrial Fibrillation After Isolated Coronary Artery Bypass Graft Surgery and Long-Term Survival

Background—The advancing age and generally increasing risk profile of patients receiving isolated coronary artery bypass graft (CABG) surgery is expected to raise incidence of new-onset postoperative atrial fibrillation (AFIB) resulting in potentially higher risk of adverse outcomes. In the early postoperative course, new-onset post-CABG AFIB is considered relatively easy to treat and is believed to have little impact on patients’ long-term outcome. However, little has been done to determine the effect of new-onset post-CABG AFIB on long-term survival, and this relationship is unclear. Methods and Results—Survival was assessed in a cohort of 6899 consecutive patients without preoperative AFIB who underwent isolated CABG at Baylor University Medical Center, Dallas, Tex, between January 1, 1997 and December 31, 2006; patients who died during CABG were excluded. Ten-year unadjusted survival was 52.3% (48.4%, 56.0%) for patients with new-onset postoperative AFIB and 69.4% (67.3%, 71.4%) for patients without it. A propensity-adjusted model controlling for risk factors identified by the Society of Thoracic Surgeons and other clinical/nonclinical details was used to investigate the association between new-onset AFIB post-CABG and long-term survival. After adjustment, new-onset AFIB post-CABG was significantly associated (hazard ratio, 1.29; 95% CI, 1.16, 1.45) with increased risk of death. Conclusions—This study provides evidence that new-onset post-CABG AFIB is significantly associated with increased long-term risk of mortality independent of patient preoperative severity. After controlling for a comprehensive array of risk factors associated with post-CABG adverse outcomes, risk of long-term mortality in patients that developed new-onset post-CABG AFIB was 29% higher than in patients without it.

Using Bland-Altman to assess agreement between two medical devices--don't forget the confidence intervals!

The limits of agreement approach of Bland and Altman is by far the most popular method for investigating statistical agreement between two measurement devices. This work presents the dangers of relying exclusively on the limits of agreement alone and argues that authors should always provide confidence intervals to assess the variability in the estimated limits.

Comparing long-term survival between patients undergoing off-pump and on-pump coronary artery bypass graft operations.

BACKGROUND As the population of the United States and Western Europe ages, the number of patients undergoing isolated coronary artery bypass grafting (CABG) for revascularization can be expected to increase. This study investigated long-term survival in patients undergoing off-pump vs on-pump CABG. METHODS Survival was assessed in 8081 consecutive patients who underwent isolated CABG (732 received off-pump) between January 1, 1997, and December 31, 2008. A propensity-adjusted model controlling for preoperative risk factors identified by the Society of Thoracic Surgeons and other preoperative clinical and nonclinical details was used to assess adjusted long-term mortality differences between off-pump and on-pump CABG. RESULTS Ten-year unadjusted survival was 54.7% (95% confidence interval, 47.2% to 61.6%) in off-pump CABG patients and 62.3% (95% confidence interval 60.9% to 63.8%) in on-pump CABG patients. The log-rank test (p=0.012) indicated a significantly higher risk of death in off-pump CABG patients. After adjustment, the risk of death remained significantly higher in the off-pump CABG patients (hazard ratio, 1.18; 95% confidence interval, 1.02 to 1.38). The adjusted association regarding off-pump learning curve and survival was assessed separately and was not statistically significant (p=0.774), further validating our findings regarding off-pump CABG. CONCLUSIONS After controlling for preoperative severity of disease and other possible confounders, the risk of long-term mortality in patients undergoing off-pump CABG is significantly higher than in those undergoing on-pump CABG. For multivessel coronary disease, on-pump CABG might be preferable to off-pump CABG given that it may achieve a more complete and durable revascularization.

Relation of Obesity to Atrial Fibrillation After Isolated Coronary Artery Bypass Grafting

The impact of obesity on risk of atrial fibrillation (AF) after coronary artery bypass grafting (CABG) is poorly understood. This study was performed to investigate the relation between body mass index (BMI; kilograms per square meter) or body surface area (BSA; square meters) and AF after CABG. Postoperative AF was assessed in a cohort of 7,027 consecutive patients without preoperative AF undergoing isolated CABG at Baylor University Medical Center from January 1, 1997 to December 31, 2006. Two propensity-adjusted models controlling for risk factors identified by the Society of Thoracic Surgeons and other clinical/nonclinical details were used. After adjustment, BMI and BSA (modeled using smoothing techniques to avoid categorization) were strongly associated (p <0.0001) with postoperative AF. Although evidence existed that gender was associated with AF (p <0.0001 and p = 0.1088 for BSA and BMI models, respectively), there was no indication that the effect of BMI or BSA on postoperative AF varied by gender. In conclusion, this study demonstrates that increased BMI and BSA are associated with a higher risk of AF after CABG and that risk for men is higher for the entire BSA spectrum and for extreme values of BMI.

Erratum to: The importance of using the correct bounds on the Bland–Altman limits of agreement when multiple measurements are recorded per patient

The limits of agreement originally derived by Bland and Altman (Lancet i:307–310, 1986) are the most commonly used method for investigating statistical agreement between two medical devices. Bland and Altman describe a confidence interval for these limits that should always accompany the limits themselves. However, this interval presumes that the recorded differences between the two devices in question are independent. This is a reasonable assumption when only one measurement is recorded per device per patient. Bland and Altman (StatMethods Med Res 8:135–160, 1999) subsequently derived bounds for the more general case where multiple observations are recorded within each patient. Unfortunately, in practice, the bounds assuming independence are typically reported when in fact the repeated measures bounds are more appropriate. This communication illustrates the dangers of using the “original” (independence-based) bounds derived in Bland and Altman (Lancet i:307–310, 1986) in the presence of repeated measures per patient.

A hospital-randomized controlled trial of an educational quality improvement intervention in rural and small community hospitals in Texas following implementation of information technology.

Rural and small community hospitals typically have few resources and little experience with quality improvement (QI) and, on average, demonstrate poorer quality of care than larger facilities. Formalized QI education shows promise in improving quality, but little is known about its effect in rural and small community hospitals. The authors describe a randomized controlled trial assigning 47 rural and small community Texas hospitals to such a program (n = 23) or to the control group (n = 24), following provision of a Web-based quality benchmarking and case review tool. Centers for Medicare and Medicaid Services Core Measures composite scores for congestive heart failure (CHF) and community-acquired pneumonia (CAP), using Texas Medical Foundation data collected via the QualityNet Exchange system, are compared for the groups, for 2 years postintervention. Given the estimated baseline rates for the CHF (68%) and CAP (66%) composites, the cohort enables the detection of 14% and 11% differences (α = .05; power = 0.8), respectively. (Am J Med Qual 2007;22:418-427)

A Note on Confidence Intervals for a Linear Function of Poisson Rates

We consider three interval estimators for linear functions of Poisson rates: a Wald interval, a t interval with Satterthwaites degrees of freedom, and a Bayes interval using noninformative priors. The differences in these intervals are illustrated using data from the Crash Records Bureau of the Texas Department of Public Safety. We then investigate the relative performance of these intervals via a simulation study. This study demonstrates that the Wald interval performs poorly when expected counts are less than 5, while the interval based on the noninformative prior performs best. It also shows that the Bayes interval and the interval based on the t distribution perform comparably well for more moderate expected counts.

Immediate Open Repair vs Surveillance in Patients with Small Abdominal Aortic Aneurysms: Survival Differences by Aneurysm Size

OBJECTIVE To assess whether survival differences exist between patients undergoing immediate open repair vs surveillance with selective repair for 4.0- to 5.4-cm abdominal aortic aneurysms (AAAs) and whether these differences vary by diameter, within sexes, or overall. PATIENTS AND METHODS The study cohort included 2226 patients randomized to immediate repair or surveillance for the UK Small Aneurysm Trial (September 1, 1991, through July 31, 1998; follow-up, 2.6-6.9 years) or the Aneurysm Detection and Management trial (August 1, 1992, through July 31, 2000; follow-up, 3.5-8.0 years). Survival differences were assessed with proportional hazard models, adjusted for a comprehensive array of clinical and nonclinical risk factors. Interaction between treatment and AAA size was added to the model to assess whether the effect of immediate open repair vs surveillance varied by AAA size. RESULTS The adjusted analysis revealed no statistically significant survival difference between immediate open repair and surveillance patients (hazard ratio [HR], 0.99; 95% CI, 0.83-1.18; mean follow-up time, 1921 days for both study groups). This lack of treatment effect persisted when men (HR, 1.01; 95% CI, 0.84-1.21) and women (HR, 0.96; 95% CI, 0.49-1.86) were examined separately and did not vary by AAA size (P=.39 for the entire cohort and P=.24 for women). CONCLUSION Immediate open repair offered no significant survival benefit, even in patients with the largest AAAs and highest risk of rupture. Because recent trials failed to find a survival benefit of immediate endovascular repair over surveillance for small asymptomatic AAAs, our findings suggest that the gray area of first-line management for these patients should be resolved in favor of surveillance.

USING A PREDICTION APPROACH TO ASSESS AGREEMENT BETWEEN TWO CONTINUOUS MEASUREMENTS

The problem of assessing agreement between two devices occurs with great frequency in the medical literature. If it can be demonstrated that a new device agrees sufficiently with a device currently in use, then the new device can be approved for general use. This work discusses how a prediction interval can be used to estimate the whether a future difference between two devices will be within acceptable limits with reasonable confidence. The method is illustrated with an example involving measurements of peak expiratory flow.